1. Field of the Invention
The present invention relates to ubiquitin-lytic peptide fusion gene constructs with enhanced stability and gene expression, ubiquitin-lytic peptide fusion protein products, and methods of making and using the same.
2. Description of Related Art
Naturally occurring lytic peptides play an important if not critical role as immunological agents in insects and have some, albeit secondary, defense functions in a range of other animals. The function of these peptides is to destroy prokaryotic and other non-host cells by disrupting the cell membrane and promoting cell lysis. Common features of these naturally occurring lytic peptides include an overall basic charge, a small size (23-39 amino acid residues), and the ability to form amphipathic .alpha.-helices or .beta.-pleated sheets. Several types of lytic peptides have been identified: cecropins (described in U.S. Pat. Nos. 4,355,104 and 4,520,016 to Hultmark et al.), defensins, sarcotoxins, melittin, and magainins (described in U.S. Pat. No. 4,810,777 to Zasloff). Each of these peptide types is distinguished by sequence and secondary structure characteristics.
Several hypotheses have been suggested for the mechanism of action of the lytic peptides: disruption of the membrane lipid bilayer by the amphipathic .alpha.-helix portion of the lytic peptide; lytic peptide formation of ion channels, which results in osmotically induced cytolysis; lytic peptide promotion of protein aggregation, which results in ion channel formation; and lytic peptide-induced release of phospholipids. Whatever the mechanism of lytic peptide-induced membrane damage, an ordered secondary conformation such as an amphipathic .alpha.-helix and positive charge density are features that appear to participate in the function of the lytic peptides.
Active synthetic analogs of naturally occurring lytic peptides have been produced and tested in vitro against a variety of prokaryotic and eukaryotic cell types (see for example Arrowood, M. J., et al., J. Protozool. 38: 161s [1991]; Jaynes, J. M., et al., FASEB J. 2: 2878 [1988]), including: gram positive and gram negative bacteria, fungi, yeast, protozoa, envelope viruses, virus-infected eukaryotic cells, and neoplastic or transformed mammalian cells. The results from these studies indicate that many of the synthetic lytic peptide analogs have similar or higher levels of lytic activity for many different types of cells, compared to the naturally occurring forms. In addition, the peptide concentration required to lyse microbial pathogens such as protozoans, yeast, and bacteria does not lyse normal mammalian cells. However, because previous work demonstrates that absolute sequence is not important as long as positive charge and amphipathy are preserved, the level of activity for a given synthetic peptide is difficult to predict.
The specificity of the lytic action also depends upon the concentration of the peptide and the type of membrane with which it interacts. Jaynes, J. M. et al., Peptide Research 2: 157 (1989) discuss the altered cytoskeletal characteristics of transformed or neoplastic mammalian cells that make them susceptible to lysis by the peptides. In these experiments, normal, human non-transformed cells remained unaffected at a given peptide concentration while transformed cells were lysed; however, when normal cells were treated with the cytoskeletal inhibitors cytochalasin D or colchicine, sensitivity to lysis increased. The experiments show that the action of lytic peptides on normal mammalian cells is limited. This resistance to lysis was most probably due to the well-developed cytoskeletal network of normal cells. In contrast, transformed cell lines which have well-known cytoskeletal deficiencies were sensitive to lysis. Because of differences in cellular sensitivity to lysis, lytic peptide concentration can be manipulated to effect lysis of one cell type but not another at the same locus.
Synthetic lytic peptide analogs can also act as agents of eukaryotic cell proliferation. Peptides that promote lysis of transformed cells will, at lower concentrations, promote cell proliferation in some cell types. This stimulatory activity is thought to depend on the channel-forming capability of the peptides, which somehow stimulates nutrient uptake, calcium influx or metabolite release, thereby stimulating cell proliferation (see Jaynes, J. M., Drug News & Perspectives 3: 69 [1990]; and Reed, W. A. et al., Molecular Reproduction and Development 31: 106 [1992]). Thus, at a given concentration, these peptides stimulate or create channels that can be beneficial to the normal mammalian cell in a benign environment where it is not important to exclude toxic compounds.
The synthetic lytic peptide analogs typically contain as few as 12 and as many as 40 amino acid residues. A phenylalanine residue is often positioned at the amino terminus of the protein to provide an aromatic moiety analogous to the tryptophan residue located near the amino terminus of natural cecropins and a UV-absorbing moiety with which to monitor the purification of the synthetic peptide. The basis for the design of these lytic peptide analogs is that a peptide of minimal length, having an amphipathic .alpha.-helical structural or a .beta.-pleated sheet motif, and overall positive charge density effects lytic activity.
Plant disease is one of the leading causes of crop loss in the world and is estimated to cause up to one third of total crop loss worldwide; for example, in the potato losses associated with bacterial disease are as high as 25% of worldwide production. Additionally, the cultivation of a few species of plants in a concentrated area exacerbates the spread of disease. Recent advances in genetic engineering have lead to the development of plants with disease resistant phenotypes based on the expression of recombinant DNA molecules. Transgenic tobacco plants were engineered with both a wound inducible PiII promoter and a constitutive 35S promoter to express two lytic peptides (SHIVA-1 and SB-37) with bacteriolytic activity. The SHIVA-1 plant demonstrated enhanced resistance to bacterial wilt caused by injection by Pseudomonas solanacearum (Jaynes, J. M., et al., Plant Science 89: 43 (1993); Destefano-Beltran, L., et al., Biotechnology in Plant Disease Control, pp. 175-189, Wiley-Liss (1993). Thus lytic peptides have valuable uses as anti-phytopathogenic agents. However, chemical synthesis of these lytic peptides is very expensive. Therefore, alternate, more economical and efficient methods of synthesis are needed, such as in vivo synthesis in host cells using recombinant DNA methods.
Recombinant DNA molecules are produced by sub-cloning genes into plasmids using a bacterial host intermediate. In principle this technique is straightforward. However, any sequence that interferes with bacterial growth through replication or production of products toxic to the bacteria, such a lytic peptides, are difficult to clone. Often, host bacterial cells containing mutated forms of the DNA sequences encoding toxic products will be selected. These mutations can result in either decreased expression or production of an inactive product. Bacteria will even insert mutations that prevent expression of a potentially toxic product in cloned genes controlled by a eukaryotic promoter that is not active in prokaryotes. The effect of this selection of mutated species leads to an inability to isolate sub-clones containing a non-mutated gene of choice. Thus, some sub-cloned genes are unstable in their bacterial hosts, although this instability can only be shown empirically. The bacteriolytic activity of the lytic peptides is an obstacle to the production of stable recombinant DNA molecules that express the genes at high levels.
For example, in an attempt to sub-clone into a standard plasmid vector a gene coding for frog magainin, a natural lytic peptide, bacterial transformants contained deletion mutations in the magainin coding region. Another attempt was made to sub-clone a synthetic lytic peptide (SEQ ID NO. 98) into a standard plasmid vector (pUC19) containing the Cauliflower Mosaic Virus 35S promoter. The resulting transformants were screened by polymerase chain reaction (PCR). However, out of 30 colonies, only 2 sub-clones gave faint positive signals. These two sub-clones were sequenced. The sequence showed that one clone had a point mutation that introduced a stop codon 3/4 of the way through the lytic peptide, and the other clone had a point mutation that changed the start codon from methionine to isoleucine. Both mutations would prevent the biosynthesis of the protein. Four more clones were analyzed, and of these four, one was sub-cloned in the wrong orientation, and three others had mutations introduced into the sequence. One of these sub-clones was selected for further analysis, but it inhibited the growth of its E. coli host. Thus, the production of recombinant DNA molecules coding for lytic peptides is difficult due to the uncertainty in obtaining the correct sub-clone.
Ubiquitin is a small, highly conserved protein present in all eukaryotes. Ubiquitins are encoded by gene families that are characterized by two types of basic structures. Polyubiquitin genes contain several direct repeats of ubiquitin, and ubiquitin-ribosomal fusion genes encode a single ubiquitin unit fused to the coding region for a small ribosomal associated protein. Both of these gene types are translated as polyproteins and then are processed by an endogenous ubiquitin hydrolase present in eukaryotes to release multiple ubiquitin proteins or ubiquitin and the ribosomal associated protein. A number of ubiquitin cDNAs or genomic clones have been isolated, including plant ubiquitin cDNAS and genomic clones from the potato (Garbarino, J. and Belknap, W., Plant Molecular Biology 24: 119 (1994); Garbarino. J. et al., Plant Molecular Biology 20: 235 (1992)).
U.S. Pat. Nos. 5,093,242 and 5,132,213 to Bachmair et al. teach the use of a ubiquitin cloning vector as a method of producing specified protein amino-termini. A recombinant DNA molecule was constructed with a protein coding gene fused at its amino terminus to a ubiquitin coding gene. Due to translation as a polypeptide and cleavage by hydrolases, a protein with any amino acid at the amino terminus can be generated. The amino terminus can be used to control the metabolic stability of the protein. However, the metabolic stability of the protein is dependent on the resulting amino acid at the amino-terminus, not the generation of a translation polypeptide.
The forgoing facts suggest that although lytic peptides as a class may include species that are efficacious in destroying bacteria, neoplastic cells, fungi, virus-infected cells, and protozoa, this lytic characteristic also decreases the stability of sub-cloned lytic peptides in host cells. This decreased stability hinders efforts to develop a more economical and efficient means of synthesizing lytic peptides.
It would therefore be a significant advance in the art, and is correspondingly an object of the present invention to develop a method of sub-cloning nucleotide sequences coding for lytic peptides into expression vectors, providing gene constructs with enhanced stability and gene expression and reduced toxicity.